基于深度学习和钻孔图像岩体RQD智能计算

李东黎; 刘兴宇; 张占荣; 葛云峰; 李炜; 张子龙

doi:10.19509/j.cnki.dzkq.tb20250114

基于深度学习和钻孔图像岩体RQD智能计算

doi: 10.19509/j.cnki.dzkq.tb20250114

李东黎^1,,
刘兴宇²,
张占荣¹,
葛云峰^{2, 3, ,},
李炜¹,
张子龙²

1.
中铁第四勘察设计院集团有限公司，武汉 430063
2.
中国地质大学（武汉）工程学院，武汉 430074
3.
新疆工业学院自然资源科技学院，新疆和田 848000

基金项目: 中国铁建股份有限公司科技重大专项（2022-A02）；中国铁建股份有限公司重大专项（2024-W01）

详细信息

作者简介:
李东黎：E-mail：985203319@qq.com

通讯作者:
E-mail：geyunfeng@cug.edu.cn

计量
- 文章访问数: 490
- PDF下载量: 46
- 被引次数: 0
出版历程
- 收稿日期: 2025-03-11
- 录用日期: 2025-06-26
- 修回日期: 2025-06-26
- 网络出版日期: 2025-06-26

Intelligent rock mass RQD computation based on deep learning and borehole imagery

1.
China Railway Fourth Survey and Design Institute Group Co., Ltd., Wuhan 430063, China
2.
Faculty of Engineering, China University of Geosciences, Wuhan 430074, China
3.
School of Natural Resources Science and Technology, Xinjiang University of Technology, Hotan Xinjiang 848000, China

More Information

Author Bio:
E-mail：985203319@qq.com

Corresponding author: E-mail：geyunfeng@cug.edu.cn

摘要

摘要:
岩石质量指标（RQD）是岩土工程中公认的并广泛运用的反映岩石完整性的重要指标，常用于岩石质量分类，同时也是评级系统的重要输入参数。传统的RQD确定方法依赖于人工岩芯测井，但通常费时费力，且易受钻进工艺、取芯质量的影响，往往不能客观获得RQD值。本研究基于深度学习算法YOLOv5提出了一种新方法，无需采集岩心，能够直接从钻孔井下电视获得的原位钻孔图像中自动识别和定位结构面，从而避免岩心取样过程中的破坏性影响，并实现RQD的智能计算。该方法首先对钻孔原始图像进行预处理构建一个具有代表性的数据集，然后采用深度学习算法训练模型，最后结合图像分析方法自动计算RQD。为了验证该方法的准确性，本研究选取了位于中国湖南省永州市某隧道工程，通过对比zk4钻孔得到的RQD智能计算值和RQD人工测量值，发现根据钻孔图像智能计算的RQD值相较现场人工对于岩芯盒的实测值平均偏高20%，平均绝对误差为9.82%。本研究提出的方法能够避免钻进和取芯过程对实际RQD造成的影响，提高了RQD数据的准确性，体现了其卓越的可靠性和有效性。
- 岩体质量 /
- 岩石质量指标(RQD) /
- 井下电视 /
- 深度学习 /
- YOLO /
- 智能计算
Abstract:
Purpose
Rock Quality Designation (RQD) serves as a fundamental index in geotechnical engineering for evaluating rock mass integrity. It is extensively applied in rock mass classification systems and serves as a critical input parameter for various engineering rating methods. Conventionally, RQD determination relies on manual logging of recovered drill cores. However, this approach is labor-intensive, time-consuming, and often sensitive to drilling techniques and core quality. Such dependencies introduce subjectivity and potential inconsistencies, ultimately limiting the objectivity and repeatability of RQD evaluation.
Method
In light of these challenges, this study proposes an innovative, non‑destructive approach utilizing deep learning. We adopt the YOLOv5 (You Only Look Once, version 5) framework to detect and localize discontinuities directly from borehole televiewer images, thereby eliminating the need for physical core extraction. First, raw televiewer imagery is preprocessed, annotated, and augmented to build a representative dataset that highlights natural fractures, bedding planes, and other geological discontinuities. Next, a YOLOv5 detector is trained on this dataset to recognize and segment discontinuities with high spatial accuracy. Finally, the model output is post-processed to compute RQD automatically, by quantifying the proportion of continuous rock segments exceeding the standard 10 cm threshold.
Results
To assess the method’s performance, a case study was conducted on borehole zk4, part of a tunnel project in Yongzhou City, Hunan Province, China. Intelligent RQD values derived from the televiewer images were compared with conventional RQD measurements obtained from core boxes in the field. The results indicate that the automated approach tends to overestimate RQD by around 20 % relative to manual measurements, with a mean absolute error of 9.82 %. Despite this systematic bias, the spatial trend of RQD variation identified by the intelligent method closely matches that of in-situ wave velocity profiles, suggesting that the technique accurately captures relative changes in rock mass properties along the borehole.
Conclusion
Overall, the proposed YOLOv5‑based workflow effectively reduces the influence of drilling-induced biases and core extraction artifacts on RQD estimation. By enabling rapid, repeatable, and objective computation of RQD directly from borehole images, the method enhances both efficiency and reliability of rock quality assessment. Future work will explore calibration strategies to adjust for systematic deviations and integration with complementary geophysical datasets. This approach demonstrates significant potential to digitalize geotechnical investigation processes, streamline tunnel engineering workflows, and advance rock mass characterization in a more robust and data-driven manner.
- Key wrods: rock mass quality /
- RQD /
- borehole televiewer imagery /
- deep learning /
- YOLO /
- intelligent computation

HTML全文

图 1 整体研究流程框架图

a. 训练集边界框损失；b. 验证集边界框损失；c. 训练集置信度损失；d. 验证集置信度损失；e. 精确率；f. 平均精度均值，IoU=0.5；g. 召回率；h. IoU阈值从 0.5 一直提高到 0.95（步长0.05）时的平均精度均值。Loss. 损失率；IoU =交集面积/并集面积，交集面积指预测框和真实框重叠的面积，并集面积指预测框和真实框总共占用的所有面积，其值域在 0～1 之间；results. 原始真实数据，代表模型在每一个训练轮次实际记录下来的确切数值；smooth. 平滑后的趋势线，为过滤掉原始数据中剧烈的、局部的上下震荡噪声，自动对原始数据进行平滑处理，展现整体的走向和趋势；Backbone. 主干网络；Neck. 颈部网络；CBL. 最基础的卷积模块；Concat. 拼接合并；CSP1_X，CSP2_X. 跨阶段局部网络模块；Focus. 一种切片（slice）操作模块；SPP. 空间金字塔池化；crack 0.71. 模型为每一个检测到的目标生成的预测标签，crack指结构面，0.71指模型对当前预测结果的置信程度；D. 整张钻孔图像的实际物理长度；D₁₁，D₁₂，D₁₃，D₁₄. 每段相邻结构面间的实际物理距离；下同

Figure 1. Framework of the overall research process

下载: 全尺寸图片幻灯片

图 2 裁剪过程示意图

Figure 2. Schematic representation of the cropping procedure

下载: 全尺寸图片幻灯片

图 3 结构面标注过程示意图

Figure 3. Schematic diagram of discontinuity labeling

下载: 全尺寸图片幻灯片

图 4 数据集中各种岩性标注图

a.花岗岩标注；b.砂岩标注；c.凝灰岩标注；d.板岩标注

Figure 4. Examples of various lithological labels in the dataset

下载: 全尺寸图片幻灯片

图 5 样本数据增强

a.原始图片；b.水平翻转；c.垂直翻转；d.水平垂直翻转；e.旋转90°；f.旋转90°后水平翻转；g.旋转90°垂直翻转；h.旋转90°后水平垂直翻转

Figure 5. Sample data augmentation

下载: 全尺寸图片幻灯片

图 6 YOLO5识别定位结构面原理（NMS. 非极大值抑制）

Figure 6. YOLOv5 principle for discontinuity identification and localization

下载: 全尺寸图片幻灯片

图 7 YOLOv5网络结构图

蓝色线框内为整个网络结构的子模块细节拆解图；Resunit. 残差单元

Figure 7. YOLOv5 network architecture

下载: 全尺寸图片幻灯片

图 8 训练结果曲线

F1. 综合考量精度Precision和召回率Recall的一个核心评价指标，其计算公式为：$F1=2\times \dfrac{Precision \times Recall}{Precision+Recall} $

Figure 8. Training Performance Curves

下载: 全尺寸图片幻灯片

图 9 结构面识别定位结果（矩形框内为识别和定位的结构面）

a～c. 密集结构面识别结果；d～f. 有阴影干扰结构面识别结果；g～i. 水下结构面识别结果；j. 重叠结构面识别结果；k. 大面积结构面识别结果；i. 破碎带识别结果

Figure 9. Results of discontinuity identification and localization

下载: 全尺寸图片幻灯片

图 10 精度−召回率曲线

Figure 10. Precision-recall curve

下载: 全尺寸图片幻灯片

图 11 RQD计算可视化图

Figure 11. RQD calculation visualization

下载: 全尺寸图片幻灯片

图 12 获取钻孔图像数据图

Figure 12. Acquisition of borehole image data

下载: 全尺寸图片幻灯片

图 13 图12b，c中1 m段钻孔图像结构面识别结果

Figure 13. Discountinuity identification results for 1-m sections of the borehole image in Figure

下载: 全尺寸图片幻灯片

图 14 人工测量与智能计算RQD误差分析图

Figure 14. RQD error analysis diagram of manual measurement and intelligent calculation

下载: 全尺寸图片幻灯片

表 1 数据源及信息

Table 1. Data sources and descriptive statistics

项目地点	钻孔编号	深度/m	数据图片数量/张	岩石性质	典型钻孔图片
重庆市	ZK01，ZK02	99，79	130	主要由砂岩、砂质泥岩组成。颜色为灰白色、褐色
深圳市	ZK01630A，ZK01630B	56，98	95	主要由杂填土、粉质粘土、花岗岩组成。颜色为肉红色
江西省	ZK101	120	114	主要由凝灰岩组成。颜色为灰白色、深灰色
湖南省	ZK1501A，ZK8501A	77.8，45.5	88	主要由粉质黏土、板岩组成。颜色为浅灰色
湖北省	ZK001A，ZK921A	204，202.1	221	主要由灰岩组成。颜色为浅灰色

下载: 导出CSV

表 2 人工测量与智能计算RQD对比

Table 2. Comparison record of RQD between manual measurement and intelligent calculation

钻孔图像编号	起始深度/m	终止深度/m	智能计算 RQD/%	人工测量 RQD/%	RQD估计误差/%	ROD估计平均绝对误差/%
11	11	12	100	100	0	9.82
14	14	15	100	91.49	8.51
20	20	21	100	91.61	8.39
23	23	24	94	86	8
25	25	26	100	94.5	5.5
26	26	27	100	100	0
29	29	30	100	83.07	16.93
30	30	31	100	91.8	8.2
31	31	32	100	57.82	42.18
34	34	35	100	86.36	13.64
44	44	45	96	82.8	13.2
46	46	47	62.4	57.76	4.624	9.82
48	48	49	66.44	55.3	11.14
52	52	53	100	100	0
53	53	54	66.48	20.36	46.12
55	55	56	76.3	70	6.3
57	57	58	53.7	49.29	4.41
59	59	60	65.66	55.19	10.47
61	61	62	100	100	0
62	62	63	100	79.75	20.25
66	66	67	40.81	35	5.81
69	69	70	96.25	90	6.25
74	74	75	92.57	88.75	3.82
78	78	79	96	94	2
83	83	84	100	100	0

下载: 导出CSV

参考文献(42)

[1]	韩振华, 张路青, 袁广祥. 基于BQ系统的阿拉善巴彦诺日公NRG01号钻孔岩体质量评价[J]. 工程地质学报, 2019, 27(6): 1208-1215. doi: 10.13544/j.cnki.jeg.2017-183 HAN Z H, ZHANG L Q, YUAN G X. Rock mass quality assessment of borehole NRG01 in Bayannuorigong Alxa based on BQ-system[J]. Journal of Engineering Geology, 2019, 27(6): 1208-1215. (in Chinese with English abstract doi: 10.13544/j.cnki.jeg.2017-183
[2]	BARTON N, LIEN R, LUNDE J. Engineering classification of rock masses for the design of tunnel support[J]. Rock Mechanics, 1974, 6(4): 189-236. doi: 10.1007/BF01239496
[3]	DEERE D. The rock quality designation (RQD) index in practice[M]. West Conshohocken: ASTM International, 1988.
[4]	张国华, 姜晋云, 吕国磊, 等. 基于目标检测和图像分割的岩芯RQD自动生成算法[J]. 安全与环境工程, 2025, 32(1): 100-106. doi: 10.13578/j.cnki.issn.1671-1556.20231316 ZHANG G H, JIANG J Y, LYU G L, et al. Automatic computing algorithm for rock core RQD based on object detection and image segmentation[J]. Safety and Environmental Engineering, 2025, 32(1): 100-106. (in Chinese with English abstract doi: 10.13578/j.cnki.issn.1671-1556.20231316
[5]	CHEN J P, WANG Q, ZHAO H. Obtaining RQD of rock mass by sampling window method[J]. Chinese Journal of Rock Mechanics and Engineering, 2004, 23(9): 1491-1495.
[6]	ZHANG W, WANG Q, CHEN J P, et al. Determination of the optimal threshold and length measurements for RQD calculations[J]. International Journal of Rock Mechanics and Mining Sciences, 2012, 51: 1-12. doi: 10.1016/j.ijrmms.2012.02.005
[7]	李清波, 杜朋召. 基于边缘阈值分割的钻孔图像RQD自动分析方法研究[J]. 岩土工程学报, 2020, 42(11): 2153-2160. LI Q B, DU P Z. Automatic RQD analysis method based on information recognition of borehole images[J]. Chinese Journal of Geotechnical Engineering, 2020, 42(11): 2153-2160. (in Chinese with English abstract
[8]	ADJISKI V, PANOV Z, POPOVSKI R, et al. Application of photogrammetry for determination of volumetric joint count as a measure for improved rock quality designation (RQD) index[J]. Sustainable Extraction and Processing of Raw Materials, 2021, 2(2): 12-21. doi: 10.58903/b15171911
[9]	LIU F Y, LIU Y H, YANG T H, et al. Meticulous evaluation of rock mass quality in mine engineering based on machine learning of core photos[J]. Chinese Journal of Geotechnical Engineering, 2021, 43(5): 968-974.
[10]	ZHANG Y, CHEN J Q, LI Y L, et al. Automatic estimation of RQD based on deep ensemble learning and fracture fitting[J]. Geoenergy Science and Engineering, 2023, 230: 212132. doi: 10.1016/j.geoen.2023.212132
[11]	FU D, SU C, LI X R. Automatic estimation of rock quality designation based on an improved YOLOv5[J]. Rock Mechanics and Rock Engineering, 2024, 57(4): 3043-3061.0. doi: 10.1007/s00603-023-03729-x
[12]	张航, 胡海瑞, 朱杰清, 等. 基于非接触测量楔形危岩体稳定性快速评价[J]. 地质科技通报, 2025, 44(2): 67-77. ZHANG H, HU H R, ZHU J Q, et al. Rapid stability assessment of wedge-shaped dangerous rock mass based on non-contact measurement[J]. Bulletin of Geological Science and Technology, 2025, 44(2): 67-77. (in Chinese with English abstract
[13]	李长宏, 黄永亮, 沙鹏, 等. 基于裂隙网络模拟的岩体RQD测量参数确定[J]. 科技通报, 2022, 38(4): 63-69. LI C H, HUANG Y L, SHA P, et al. Determination of measurement parameters for rock mass RQD based on fracture network simulation[j]. Bulletin of Science and Technology, 2022, 38(4): 63-69. (in Chinese with English abstract
[14]	葛云峰, 钟鹏, 唐辉明, 等. 基于钻孔图像的岩体结构面几何信息智能测量[J]. 岩土力学, 2019, 40(11): 4467-4476. GE Y F, ZHONG P, TANG H M, et al. Intelligent measurement on geometric information of rock discontinuities based on borehole image[J]. Rock and Soil Mechanics, 2019, 40(11): 4467-4476. (in Chinese with English abstract
[15]	胡瀚. 基于数字近景摄影测量的岩体RQD获取[D]. 长春: 吉林大学, 2018. HU H. Rock mass RQD acquisition based on digital close-range photogrammetry[D]. Changchun: Jilin University, 2018. (in Chinese with English abstract
[16]	GE Y F, DU B, TANG H M, et al. Rock joint detection from borehole imaging logs based on grey-level co-occurrence matrix and Canny edge detector[J]. Quarterly Journal of Engineering Geology and Hydrogeology, 2022, 55: qjegh2021-qjegh2016. doi: 10.1144/qjegh2021-016
[17]	李炜, 刘耕, 葛云峰, 等. 基于深度学习的钻孔图像岩体结构面识别[J]. 应用基础与工程科学学报, 2024, 32(3): 702-720. LI W, LIU G, GE Y F, et al. Detection of rock discontinuities in borehole images based on a deep learning method[J]. Journal of Basic Science and Engineering, 2024, 32(3): 702-720. (in Chinese with English abstract
[18]	GE Y F, CHEN H Z, ZHAO B B, et al. A comparison of five methods in landslide susceptibility assessment: A case study from the 330-kV transmission line in Gansu region, China[J]. Environmental Earth Sciences, 2018, 77(19): 662. doi: 10.1007/s12665-018-7814-7
[19]	付浩辰, 白堂博, 许贵阳, 等. 基于改进YOLOv8的轨道板裂缝检测算法[J]. 北京交通大学学报, 2024, 48(6): 133-143. FU H C, BAI T B, XU G Y, et al. Crack detection in track slabs based on an improved YOLOv8 algorithm[J]. Journal of Beijing Jiaotong University, 2024, 48(6): 133-143. (in Chinese with English abstract
[20]	张明振, 段江忠, 梁肇伟, 等. 基于改进YOLO-V5 算法的烟火检测方法[J]. 中国安全科学学报, 2024, 34(5): 155-161. doi: 10.16265/j.cnki.issn1003-3033.2024.05.1050 ZHANG M Z, DUAN J Z, LIANG Z W, et al. Firework detection method based on improved YOLO-V5 algorithm[J]. China Safety Science Journal, 2024, 34(5): 155-161. (in Chinese with English abstract doi: 10.16265/j.cnki.issn1003-3033.2024.05.1050
[21]	米增, 连哲. 面向通用目标检测的YOLO方法研究综述[J]. 计算机工程与应用, 2024, 60(21): 38-54. MI Z, LIAN Z. Review of YOLO methods for universal object detection[J]. Computer Engineering and Applications, 2024, 60(21): 38-54. (in Chinese with English abstract
[22]	霍俊杰, 黄润秋, 董秀军, 等. 3D激光扫描与岩体结构精细测量方法比较研究: 以锦屏Ⅰ级水电站为例[J]. 湖南科技大学学报(自然科学版), 2011, 26(3): 39-44. HUO J J, HUANG R Q, DONG X J, et al. Comparative study of 3D laser scanning and fine measurement methods of rock mass structure: A case study of Jinping Ⅰ Hydropower Station[J]. Journal of Hunan University of Science & Technology, 2011, 26(3): 39-44. (in Chinese with English abstract
[23]	王焰新, 曹海龙, 谢先军, 等. 基于树的机器学习方法预测地质成因劣质地下水空间分布[J]. 安全与环境工程, 2022, 29(5): 58-64. doi: 10.13578/j.cnki.issn.1671-1556.20220853 WANG Y X, CAO H L, XIE X J, et al. Prediction of spatial distribution of geogenic poor groundwater based on tree-based machine learning method[J]. Safety and Environmental Engineering, 2022, 29(5): 58-64. (in Chinese with English abstract doi: 10.13578/j.cnki.issn.1671-1556.20220853
[24]	汪进超, 王川婴, 胡胜, 等. 孔壁钻孔图像的结构面参数提取方法研究[J]. 岩土力学, 2017, 38(10): 3074-3080. WANG J C, WANG C Y, HU S, et al. A new method for extraction of parameters of structural surface in borehole images[J]. Rock and Soil Mechanics, 2017, 38(10): 3074-3080. (in Chinese with English abstract
[25]	徐伟, 胡新丽, 黄磊, 等. 结构面三维网络模拟计算RQD及精度对比研究[J]. 岩石力学与工程学报, 2012, 31(4): 822-833. XU W, HU X L, HUANG L, et al. Research on RQD of rock mass calculated by three-dimensional discontinuity network simulation method and its accuracy comparison[J]. Chinese Journal of Rock Mechanics and Engineering, 2012, 31(4): 822-833. (in Chinese with English abstract
[26]	蒋小伟, 万力, 王旭升, 等. 利用RQD估算岩体不同深度的平均渗透系数和平均变形模量[J]. 岩土力学, 2009, 30(10): 3163-3167. JIANG X W, WAN L, WANG X S, et al. Estimation of average permeability and average deformation modulus of rock mass at different depths using RQD[J]. Rock and Soil Mechanics, 2009, 30(10): 3163-3167. (in Chinese with English abstract
[27]	王川婴, LAWK Tim. 钻孔摄像技术的发展与现状[J]. 岩石力学与工程学报, 2005, 24(19): 3440-3448. doi: 10.3321/j.issn:1000-6915.2005.19.006 WANG C Y, LAWK T. Review of borehole camera technology[J]. Chinese Journal of Rock Mechanics and Engineering, 2005, 24(19): 3440-3448. (in Chinese with English abstract doi: 10.3321/j.issn:1000-6915.2005.19.006
[28]	DEERE D U, HENDRON A J, PATTON F D, et al. Design of surface and near-surface construction in rock[C]//American Rock Mechanics Association. US Rock Mechanics/Geomechanics Symposium. [S. l. ]: [s. n. ], 1966: ARMA-66-0237.
[29]	赵兴东, 王宏宇, 王小兵, 等. 基于智能自学习模型识别岩芯的RQD标定方法[J]. 矿业研究与开发, 2023, 43(12): 159-165. ZHAO X D, WANG H Y, WANG X B, et al. RQD calibration method for core recognition based on intelligent self-learning model[J]. Mining Research and Development, 2023, 43(12): 159-165. (in Chinese with English abstract
[30]	ALZUBAIDI F, MOSTAGHIMI P, SI G Y, et al. Automated rock quality designation using convolutional neural networks[J]. Rock Mechanics and Rock Engineering, 2022, 55(6): 3719-3734. doi: 10.1007/s00603-022-02805-y
[31]	涂美义, 袁世宇, 陈江军, 等. 不同降雨工况下的矿山修复工程边坡稳定性评价[J]. 地质科技通报, 2024, 43(6): 63-77. doi: 10.19509/j.cnki.dzkq.tb20230527 TU M Y, YUAN S Y, CHEN J J, et al. Slope stability evaluation of mine rehabilitation project under different rainfall conditions[J]. Bulletin of Geological Science and Technology, 2024, 43(6): 63-77. (in Chinese with English abstract doi: 10.19509/j.cnki.dzkq.tb20230527
[32]	朱志明, 欧阳继胜, 张子龙, 等. 缓倾顺层岩质滑坡机理研究: 以广元市苍溪县中梁村滑坡为例[J]. 安全与环境工程, 2025, 32(1): 233-243. doi: 10.13578/j.cnki.issn.1671-1556.20240319 ZHU Z M, OUYANG J S, ZHANG Z L, et al. Mechanism of gently dipping bedding rock landslide: A case study of Zhongliang Village landslide in Cangxi County, Guangyuan City[J]. Safety and Environmental Engineering, 2025, 32(1): 233-243. (in Chinese with English abstract doi: 10.13578/j.cnki.issn.1671-1556.20240319
[33]	CHU H H, LONG L Z, GUO J J, et al. Implicit function-based continuous representation for meticulous segmentation of cracks from high-resolution images[J]. Computer-Aided Civil and Infrastructure Engineering, 2024, 39(4): 539-558. doi: 10.1111/mice.13052
[34]	XU S, MA J, LIANG R Y, et al. Intelligent recognition of drill cores and automatic RQD analytics based on deep learning[J]. Acta Geotechnica, 2023, 18(11): 6027-6050. doi: 10.1007/s11440-023-02011-2
[35]	PASZKE A, GROSS S, MASSA F, et al. PyTorch: An imperative style, high-performance deep learning library[EB/OL]. (2019)[2026-04-09]. https://arxiv.org/abs/1912.01703, arXiv: 1912.01703.
[36]	秦四清. 分维Df与RQD的关系模型[J]. 水文地质工程地质, 1995, 22(1): 1-2. QIN S Q. The relation model between fractal dimension Df and RQD[J]. Hydrogeology & Engineering Geology, 1995, 22(1): 1-2. (in Chinese with English abstract
[37]	赵金海, 朱伟龙, 孙文斌, 等. 基于钻孔数据的断层带结构数值模型构建及应用探索[J]. 煤炭学报, 2025, 50(3): 1458-1472. doi: 10.13225/j.cnki.jccs.2024.1253 ZHAO J H, ZHU W L, SUN W B, et al. Construction and application exploration of numerical model of fault zone structure based on borehole data[J]. Journal of China Coal Society, 2025, 50(3): 1458-1472. (in Chinese with English abstract doi: 10.13225/j.cnki.jccs.2024.1253
[38]	李志, 陈宁生, 侯儒宁, 等. 基于机器学习的伊犁河谷黄土区泥石流易发性评估[J]. 中国地质灾害与防治学报, 2024, 35(3): 129-140. LI Z, CHEN N S, HOU R N, et al. Susceptibility assessment of debris flow disaster based on machine learning models in the loess area along Yili Valley[J]. The Chinese Journal of Geological Hazard and Control, 2024, 35(3): 129-140. (in Chinese with English abstract
[39]	YU S H, WONG L N Y. A K-Net-based deep learning framework for automatic rock quality designation estimation[J]. Computer-Aided Civil and Infrastructure Engineering, 2025, 40(16): 2287-2303. doi: 10.1111/mice.13386
[40]	冯谕, 曾怀恩, 涂鹏飞. 遗传算法下的滑坡蠕滑位移预测模型研究[J]. 中国地质灾害与防治学报, 2024, 35(1): 82-91. doi: 10.16031/j.cnki.issn.1003-8035.202209038 FENG Y, ZENG H E, TU P F. Research on prediction model of landslide creep displacement on genetic algorithm[J]. The Chinese Journal of Geological Hazard and Control, 2024, 35(1): 82-91. (in Chinese with English abstract doi: 10.16031/j.cnki.issn.1003-8035.202209038
[41]	容富. 基于广义RQD_t的岩石质量空间分布规律研究[J]. 工程地质学报, 2016, 24(2): 246-251. RONG F. Study of spatial distributing of rock mass quality using generalized RQD_t[J]. Journal of Engineering Geology, 2016, 24(2): 246-251. (in Chinese with English abstract
[42]	王少勇, 吴爱祥, 韩斌, 等. 自然崩落法矿岩可崩性模糊物元评价方法[J]. 岩石力学与工程学报, 2014, 33(6): 1241-1247. WANG S Y, WU A X, HAN B, et al. Fuzzy matter-element evaluation of ore-rock cavability in block caving method[J]. Chinese Journal of Rock Mechanics and Engineering, 2014, 33(6): 1241-1247. (in Chinese with English abstract